Ancient feeding, ecology and extinction of Pleistocene horses from the Pampean Region, Argentina

Begoña Sánchez¹, José Luis Prado² and María Teresa Alberdi¹

¹ Museo Nacional de Ciencias Naturales, CSIC, José Gutiérrez Abascal, 2, 28006 Madrid, Spain. mcnsc2b@mncn.csic.es and malberdi@mncn.csic.es
² INCUAPA, Departamento de Arqueología, Universidad Nacional del Centro, Del Valle 5737, 7400 Olavarría, Argentina. jprado@soc.unicen.edu.ar

Abstract. To reconstruct the diet and habitat preference of fossil horses, we measured the carbon and oxygen isotope composition of 35 bone and tooth samples of Equus (Amerhippus) neogeus Lund, Hippidion principale (Lund), and Hippidion devillei (Gervais) from 10 different Pleistocene localities in the Pampean region (Argentina). To compare the three species by stratigraphic age, we divided the samples into three groups: lower Pleistocene, middle-late Pleistocene and latest Pleistocene. Samples of Hippidion devillei from the lower Pleistocene were more homogeneous, with δ¹³C values ranging between -11.73 to -9.79‰. These data indicate a diet exclusively dominated by C₃ plants. In contrast, Hippidion principale and Equus (Amerhippus) neogeus from middle-late Pleistocene showed a wide range of feeding adaptations (with a range of δ¹³C values between -12.05 to -8.08 ‰ in Hippidion and δ ¹³C values between -11.46 to -7.21 ‰ in Equus (Amerhippus)). These data seem to indicate a mixed C₃ - C₄ diet, while data from the latest Pleistocene suggest a tendency toward an exclusively C₃ diet for both species. Furthermore, the results of δ¹⁸O indicate an increase of approximately 4ºC from the early to latest Pleistocene in this area. Several nutritional hypotheses explaining latest Pleistocene extinctions are based on the assumption that extinct taxa had specialized diets. The resource partitioning preference of these species from latest Pleistocene in the Pampean region supports these hypotheses.

Resumen. Paleodieta, ecología y extinción de los caballos del Pleistoceno de la región pampeana, Argentina. Para comprender las variaciones en la dieta de los équidos, hemos analizado la composición isotópica del oxígeno y del carbono del carbonato en 35 restos fósiles de Equus (Amerhippus) neogeus Lund, Hippidion principale (Lund) e Hippidion devillei (Gervais) procedentes de 10 localidades del Pleistoceno de la Región Pampeana (Argentina). Para poder comparar las variaciones temporales de los resultados isotópicos, hemos agrupado las muestras según su procedencia estratigráfica en tres grupos: Pleistoceno inferior, Pleistoceno medio y superior y Pleistoceno tardío. Los datos indican que las muestras del Pleistoceno temprano de Hippidion devillei son las más homogéneas, con un rango de variación de los valores de δ¹³C entre -11,73 y -9,79‰. Estos resultados sugieren una dieta compuesta exclusivamente por plantas C₃. Por el contrario los resultados de las muestras de Hippidion principale y Equus (Amerhippus) neogeus del Pleistoceno medio y superior presentan un mayor rango de variación (con valores de δ ¹³C entre -12,05 y -8,08 ‰ para Hippidion y valores de δ¹³C entre -11,46 y -7,21 ‰ para Equus). Estos resultados indican una adaptación a una dieta mixta C₃-C₄. Por su parte los especímenes del Pleistoceno tardío muestran nuevamente una tendencia hacia una dieta compuesta exclusivamente por plantas de tipo C₃ para ambas especies. Además, los resultados de δ ¹⁸O registran un aumento en la temperatura de alrededor 4 ºC en el área de estudio desde el inicio del Pleistoceno hasta el final. Algunos autores sugieren que un estrés nutricional, producto de un cambio rápido en las comunidades vegetales, podría ser una de las causas que expliquen la extinción del final del Pleistoceno. La especialización que observamos en la dieta en los caballos del final del Pleistoceno podría ser una evidencia en favor de esta teoría.

Key words. Equus (Amerhippus); Hippidion; Paleodiet; Ecology; Extinction; Pleistocene; Papean region; South America.
Palabras clave. Equus (Amerhippus); Hippidion; Paleodieta; Ecología; Extinción; Pleistoceno; Región Pampeana; América del Sur.

Introduction

The Equidae family in South America is represented by two genera, Hippidion and Equus (Amerhippus). Both genera can be differentiated by their small and large forms, probably as a consequence of having adapted to similar environments. These immigrant genera dispersed into South America using two different routes. During the Plio-Pleistocene, two corridors developed that shaped the paleobiogeographic history of most of the North American mammals in South America (Webb, 1991). The most viable model postulated for the horse's dispersal process seems to indicate that the small forms utilized the Andes corridor, while larger horses dispersed through the Eastern route and through some coastal areas. The Pampean Region represents the most austral distribution for large horses that dispersed through the Eastern route, while the group of small horses that arrived as far as the southern Patagonia dispersed via the Western route. The dispersal route of each form seems to reflect an adaptive shift in their ecology (Alberdi and Prado, 1992; Alberdi et al., 1995). Recently, a similar pattern of dispersion in South America was postulated for gomphotheres (Sánchez et al., 2004).
Horses have been recorded in the Pampean Region from the late Pliocene to the latest Pleistocene (Alberdi and Prado, 1993; Prado and Alberdi, 1994). Three species are recognized: Equus (Amerhippus) neogeus Lund, 1840, Hippidion principale (Lund, 1846), and Hippidion devillei (Gervais, 1855).
In this paper we compare the results of carbon and oxygen isotopic composition from tooth, enamel, dentine and bone of these three species of equids, in order to determine the paleodiet and habitat preferences of Pleistocene horses in the Pampean Region, and show how variations in these preferences may have been related to their extinction.

Isotopic background

Previous studies have shown that the carbon isotope ratio (δ¹³C) of fossil teeth and bones can be used to obtain dietary information about extinct herbivores (De Niro and Epstein, 1978; Vogel, 1978; Sullivan and Krueger, 1981; Lee-Thorp et al., 1989, 1994; Koch et al., 1990, 1994; Quade et al., 1992; Cerling et al., 1997; MacFadden, 2000). This carbon isotope ratio is influenced by the type of plant material ingested, which is in turn influenced by the photosynthetic pathway utilized by the plants. During photosynthesis, C₃ plants in terrestrial ecosystems (trees, bushes, shrubs, forbs, and high elevation and high latitude grasses) discriminate more markedly against the heavy ¹³C isotope during fixation of CO₂ than do tropical grasses and sedges (C₄ plants). Thus, C₃ and C₄ plants have different δ¹³C values. C₃ plants have δ¹³C values of -22 per mil (‰) to -30‰, with an average of approximately -26‰, whereas C₄ plants have δ¹³ C values of -10 to -14‰, with an average of about -12‰ (Smith and Epstein, 1971; Vogel et al., 1978; Ehleringer et al., 1986, 1991; Cerling et al., 1993). Animals then incorporate carbon from food into their tooth and bone with an additional fractionation of 12 to 14‰. Mammals feeding on C₃ plants (fruit, leaves, etc.) characteristically have δ¹³C values between about -10 and -16‰, while animals that eat C₄ tropical grasses (including blades, seeds, and roots) haveδ ¹³C values between +2 and -2‰. A mixed-feeder would fall somewhere in between these two extremes (Lee-Thorp and van der Merwe, 1987; Quade et al., 1992). Hence, the relative proportions of C₃ and C₄ vegetation in an animal diet can be determined by analyzing its teeth and bone δ¹³C.
In recent decades, there has been an increasing use of oxygen and carbon isotope analyses to reconstruct paleoenvironmental and paleoclimatical conditions. In the case of homeothermic animals in general, the oxygen isotope composition of apatite depends primarily on the oxygen balance of the animal (Longinelli, 1984; Luz et al., 1984). Influxes of oxygen include ingested water (drinking water + water from plants) as well as inspired O₂ gas; oxygen is lost from the body as liquid water in urine, sweat, and feces, and as CO₂ and H₂O in respiratory gases. Factors acting on this balance are both internal (related to the physiology of the animal) and external (related to ecology and to climate). Oxygen isotope variations in large mammals depend largely on external factors such as the δ¹⁸O of ingested water. Because the oxygen isotopic composition of phosphate in mammalian bones and teeth (δ¹⁸Op) is related to that of ingested water, and ingested water comes ultimately from precipitation, the δ¹⁸O of enamel and bone phosphate can be used to infer past climatic conditions (Longinelli and Nuti, 1973; Kolodny et al., 1983; D'Angela and Longinelli, 1990; Bryant et al., 1994; Sánchez et al., 1994; Bryant and Froelich, 1995; Delgado et al., 1995; Kohn, 1996; Kohn et al., 1996, 1998).

Material and methods

Sampled area. The Pampa Plains contains extensive formations of superficial sand and loess that are known as the Pampean Formation (Teruggi, 1957). Paleomagnetic stratigraphy and radiometric data suggest that most of this formation was deposited over the last 3.3 Ma. (Schultz et al., 1998). The stratigraphy of pampean loess typically consists of superposed beds, 1-2 m thick, separated by either erosional discontinuities or palaeosols. The majority of these palaeo-aeolian features lie in areas presently supporting vegetation communities dominated by grassland. Most of the late Pleistocene and Holocene deposits were assigned to either the Luján Formation (including three members: La Chumbiada, Guerrero and Río Salado) or to the La Postrera Formation (Fidalgo et al., 1973, 1975, 1991; Dillon and Rabassa, 1985). Luján deposits are fluvial-lacustrine and those of La Postrera are eolian. The eolian deposits, which cover extensive parts of the southern Pampean Region, are comprised of sandy loess, very fine sand sheets, and dunefields (Zárate and Blasi, 1993).
At present, the Pampean plains has a subtropical climate, wich varies from humid in the east to arid in the west. This climatic pattern reflects the dominant effects of the ocean in the southern half of South America (Iriondo and García, 1993). The mean annu al temperatures is approximately 17°C and mean annual precipitation is around 800 mm.
In order to recognize the Ensenadan, Bonaerian, Lujanian and Platan ages, Cione and Tonni (1995, 1999) proposed several biozones for the Pleistocene and Holocene sedimentary rocks of the Pampean Region. Most of the Equus (Amerhippus) and Hippidion remains come from the localities assigned to these ages (figure 1).

Figure 1. A, Pleistocene to Holocene chronology and mammal biostratigraphy of Pampean Region following Cione and Tonni (1999). The Holocene is not in scale. B, Geographical distribution of analyzed samples of horses from the Pleistocene of Buenos Aires Province (Argentina). 1: Quequén Grande; 2: Arroyo Tapalqué; 3: Buenos Aires; 4: Río Luján; 5: Olivos; 6: Cantera Vialidad Provincial; 7: Magdalena; 8: Punta Hermengo; 9: Centinela del Mar; 10: Zanjón Seco; 11: Río Quequén Salado; 12: Río Salado / A, Cronología y biostratigrafía de los mamíferos del Pleistoceno al Holoceno de la Región Pampeana siguiendo a Cione y Tonni (1999). El Holoceno no está a escala. B, Distribución geográfica de las muestras analizadas de los caballos procedentes del Pleistoceno de la provincia de Buenos Aires (Argentina). 1: Quequén Grande; 2: Arroyo Tapalqué; 3: Buenos Aires; 4: Río Luján; 5: Olivos; 6: Cantera Vialidad Provincial; 7: Magdalena; 8: Punta Hermengo; 9: Centinela del Mar; 10: Zanjón Seco; 11: Río Quequén Salado; 12: Río Salado.

Samples selection. Thirty-five horse samples from 10 different South American Pleistocene localities in the Pampean Region were measured to obtain oxygen and carbon isotopic composition of bone and tooth enamel carbonate (figure 1). Fossil samples were collected from specimens stored at the Museo de La Plata (MLP) and the Museo Argentino de Ciencias Naturales "Bernardino Rivadavia" (MACN). The museum collection number and the locality of each sample are reported in table 1. In order to complete the discussion we also included previously published results by Sánchez et al. (2003) for Stegomastodon platensis Ameghino, 1888 from the Pampean Region. To compare the different specimens, we divided the samples into three groups, taking into account the age of the corresponding deposits. The three groups are represented in figure 2: (A) lower Pleistocene, (B) middle to late Pleistocene and (C) latest Pleistocene specimens.

Table 1. δ¹⁸O and δ¹³C isotopic values for skeletal elements of Hippidion devillei, Hippidion principale and Equus (Amerhippus) neogeus from Pampean Region. e = enamel; d = dentine; b = bone; t = tooth (enamel + dentine). Skeletal samples with the same specimen number correspond to the same individual. * taken from MacFadden et al. (1996, 1999) / resultados isotópicos de δ¹⁸O y δ¹³C de las muestras analizadas de Hippidion devillei, Hippidion principale y Equus (Amerhippus) neogeus de la Región Pampeana. e = esmalte; d = dentina; b = hueso; t = diente (esmalte + dentina). Las muestras con el mismo número corresponden a un mismo individuo. * datos tomados de MacFadden et al. (1996, 1999).

Figure 2. δ¹³C versus δ¹⁸O for: A, Hippidion devillei and Stegomastodon platensis from lower Pleistocene; B, Hippidion principale, Equus (Amerhippus) neogeus and Stegomastodon platensis from middle-late Pleistocene; C, Hippidion principale, Equus (Amerhippus) neogeus and Stegomastodon platensis from latest Pleistocene of Buenos Aires Province (Argentina) / δ¹³C versus δ¹⁸O para: A, Hippidion devillei y Stegomastodon platensis del Pleistoceno inferior; B, Hippidion principale, Equus (Amerhippus) neogeus y Stegomastodon platensis desde el Pleistoceno medio al superior; C, Hippidion principale, Equus (Amerhippus) neogeus y Stegomastodon platensis del Pleistoceno final de la provincia de Buenos Aires (Argentina).

Samples preparation and analysis. The samples (including bones) were finely ground in an agate mortar. The chemical pre-treatment of the samples was preformed as described in Koch et al. (1997) in order to eliminate secondary carbonate. About 40-50 mg of powdered dentine and bone samples were soaked in 2% NaOCl for three days at room temperature to oxidize the organic matter. Residues were rinsed and centrifuged five times with deionized water, and later treated with 1M acetic acid for one day to remove diagenetic carbonates. Pre-treatment of the enamel differed slightly as samples were soaked in 2% NaOCl for one day only. Carbon dioxide was obtained by reacting approximately 40-50mg of treated powder with 100% H₃PO₄ for five hours at 50ºC. This CO₂ was then isolated cryogenically in a vacuum line. Results are reported as δ = ([R_sample/R_standard]-1) x 1000, where R= ¹³C/¹²C or ¹⁸O/¹⁶O, and the standards are PDB for carbon and V-SMOW for oxygen. We applied data corrections for calcite to calculate the magnitude of the oxygen isotopic fractionation between apatite CO₂ and H₃PO₄ at 50ºC (Koch et al., 1989). Analytical precision for repeated analyses was 0.1‰ for δ¹³C and 0.2‰ for δ¹⁸O.
Statistical analyses. We performed both parametric (t-test) and nonparametric (Wilcoxon Signed-Rank) statistical tests to evaluate δ¹³C and δ¹⁸O differences in middle and late Pleistocene populations, accepting the null hypothesis of no differences among means unless p < 0.05. SPSS 11.0 software was used for the statistical analysis.

Analytical results and discussion

Previous results of carbon isotope analyses of Stegomastodon platensis from the Pampean Region suggest they had various types of feeding preferences throughout the Pampean record (Sánchez et al. 2003, 2004). Initial evaluation and analysis of δ ¹³C showed that horses also developed various different feeding patterns over the same record.
Parametric (t-test) and non-parametric (Wilcoxon) statistical tests (table 2) confirm significant differences between Equus (Amerhippus) and Hippidion δ ¹³C values (t-test = 2.58, p = 0.022; Wilcoxon, Z = -2.10, p = 0.036). These results indicate that the two genera must be considered separately throughout the record.

Table 2. Descriptive statistics for twelve compared groups of pampean horses and gomphotheres. A, Equus (Amerhippus), all specimens. B, Hippidion, all specimens. C, Hippidion devillei, all specimens. D, Hippidion principale, all specimens. E, Hippidion principale from middle-late Pleistocene. F, Hippidion principale from latest Pleistocene. G, Equus (Amerhippus) neogeus from middle-late Pleistocene. H, Equus (Amerhippus) neogeus from latest Pleistocene. I, Stegomastodon platensis, all specimens. J, Stegomastodon platensis from lower Pleistocene. K, Stegomastodon platensis from middle-late Pleistocene. L, Stegomastodon platensis from latest Pleistocene / análisis estadísticos realizados para la comparación de doce grupos de caballos y gonfoterios pampeanos. A, Equus (Amerhippus), todos los ejemplares. B, Hippidion, todos los ejemplares. C, Hippidion devillei, todos los ejemplares. D, Hippidion principale, todos los ejemplares. E, Hippidion principale del Pleistoceno medio y superior. F: Hippidion principale del Pleistoceno final. G, Equus (Amerhippus) neogeus del Pleistoceno medio y superior. H, Equus (Amerhippus) neogeus del Pleistoceno final. I, Stegomastodon platensis, todos los ejemplares. J, Stegomastodon platensis del Pleistoceno inferior. K, Stegomastodon platensis del Pleistoceno medio y superior. L, Stegomastodon platensis del Pleistoceno final.

Samples from the lower Pleistocene suggest that Hippidion devillei fed mostly on C₃ plants, while Stegomastodon platensis were mostly mixed-feeders (figure 2). Substantial differences in isotopic composition are also observed between gomphotheres and horses from the middle to the latest Pleistocene. There are also significant δ¹³C differences between Equus (Amerhippus) neogeus and Stegomastodon platensis (t-test = -2.88, p = 0.011; Wilcoxon, Z = -2.58, p = 0.010), and between Hippidion principale and Stegomastodon platensis (t-test = -3.05, p = 0.019; Wilcoxon, Z = -2.24, p = 0.025). During this time interval, Stegomastodon platensis had a mixed diet, confirmed by more homogeneous isotopic values, ranging from -9.08 to -6.09 ‰. In contrast, both horses exhibit a wide range of dietary adaptations (between -12.05 to -8.08 ‰ for Hippidion principale and between -11.50 to -7.20 ‰ for Equus (Amerhippus) neogeus), that indicate they changed from a mixed C₃ -C₄ diet to an exclusively C₃ diet (table 3).

Table 3. Results of parametric (t - Test) and non-parametric (Wilcoxon Signed Rank) test performed on 13 paired comparisons between horse and gomphothere groups from Pampean Region. For definition of groups see table 2 / test de significación paramétrico (t - Test) y no-paramétrico (Wilcoxon Signed Rank) de los resultados obtenidos de las comparaciones entre los grupos de caballos y gonfoterios de la Región Pampeana tomados dos a dos.

Samples from the latest Pleistocene show that there are no significant differences between horses (ttest= 0.35, p=0.752; Wilcoxon, Z=0.37, p=0.751) or between horses and gomphotheres (between Hippidion principale and Stegomastodon platensis (t-test = -2.61, p = 0.80; Wilcoxon, Z=-1.83, p = 0.068; and between Equus (Amerhippus) neogeus and Stegomastodon platensis (t-test =-0.82, p = 0.437; Wilcoxon, Z=-0.84, p =0.401). These results indicate that the three species were mostly C₃ grazers.
Mean δ ¹⁸O results range between 28.8‰ and 30.5‰, but there are no significant differences among the mean δ¹⁸O values for the stratigraphical intervals considered. The overall range of the total mean isotopic values for equids is approximately 1.7‰. Based on the equation for living horses (Delgado et al., 1995), the isotopic values indicate a total range of the environmental water isotopic composition is this area to be approximately 2.9‰. If we translate the data into temperatures, bearing in mind that the Dansgaard equation is calculated for northern Europe (Dansgaard, 1964), we obtain approximate temperature values and are able to compare them with current temperatures. So, the range of δ¹⁸O (H₂O) calculated for the equids of the Pampean Region would correspond to a variation of 4.2ºC in annual mean temperatures from early to latest Pleistocene. For example, current annual mean temperature range from 13.4º in Mar del Plata city (38º S), 16.4ºC for Buenos Aires city (34ºS), to 19.2ºC in Santa Fé city (32ºS). Our results indicate that the interval of mean temperature variation inferred for the Pleistocene is in the same order of variation as current temperature variation (at 6º of latitude). Sánchez and Alberdi (1996) report similar results for pampean gomphotheres.
Our results indicate that a relationship exists between the extinction of large mammals and their dietary preferences. Currently, Quaternary extinction of large mammals is explained by two main groups of accepted theories. One group of theories attributes large mammal extinction to climatic and ecological changes, while the other group holds man's hunting activities responsible.
The most widely accepted hypothesis related to climatic and ecological factors postulates that nutritional stress induced by rapid changes in plant communities may have been a primary cause of extinction (Graham and Lundelius, 1984; King and Saunders, 1984). This model implies that large mammals died off because they were specialized feeders, adapted to certain kinds of plants that may have disappeared during the Holocene. With this in mind, Guthrie (1984) hypothesized that plant diversity was greater, and the growing season longer, in the Pleistocene than in the Holocene. These changes in plant diversity and growing season may have affected the large mammal populations that were specialist feeders. In addition, Vrba (1993) suggests that herbivores might be more vulnerable to changes in climate that affect vegetation structure, while omnivores would be less affected as they are more able to adapt to different types of food and habitats.
The Pleistocene was characterized by rapid and repeated climatic fluctuations that caused the formation and retreat of massive continental glaciers in southern South America. The fact that large mammal extinctions in this region occurred at the same time as the major climatic deterioration associated with the last glacial cycle, has prompted considerable speculation about the extent to which these events are causally related (Borrero, 1984; Politis and Prado, 1990). During the latest Pleistocene to earliest Holocene in the Pampean Region, all megamammals (over 1 t) and a majority of large mammals (over 44 kg) suffered massive extinctions. Recent studies (Lessa and Fariñas, 1996; Lessa et al., 1997) based on late mammals from America, corroborate the idea that large body mass is the only factor strongly associated with the probability of extinction. Based on late mammals from the Pampean Region, Prado et al. (2001a, 2001b) analyzed the quantitative relationship between climatic changes and mammal diversity, and corroborated the vulnerability of large mammals to extinction.
The second group of theories hypothesises that large mammal extinction was a direct result of man and his hunting activities. Among this group, Politis and Gutiérrez (1998) identified three main models: the first postulated by Martin (1967, 1973, 1984), proposed that the extinction of large mammals from North America, South America and Australia was related to sudden human expansion into these continents. This "overkill" hypothesis is supported by the synchronism of extinction with the arrival of large numbers of humans to these continents and by mathematical simulation (Alroy, 2001). Recently, Cione et al. (2003) suggested that the main cause of the extinction in South America was due to human overhunting but was also influenced by climatic changes and reduction of open areas. Large mammals were adapted to the dry and cold climate and open areas that predominated in South America during Pleistocene. The periodic interglacial increases in temperature and humidity may have provoked shrinking of open areas and reduction of the biomass of mammals adapted to these habitats. The authors refer to this alternation of low and high biomass of animals from opened and closed areas as the "Broken Zig-Zag".
The archaeological record from the Pampean Region shows that Pleistocene mammals were common in Paleo-Indian sites (Miotti and Salemme, 1999). There are several archaeological sites where extinct mammals were recorded: La Moderna, Cerro La China 1, Zanjon Seco 2, Laguna Tres Reyes, Campo Laborde, Arroyo Seco 2 and Paso Otero 5. Each of these sites has special characteristics and different kinds of associations. In particular, the last two sites contain horse remains. Among the nine extinct species found in Arroyo Seco 2, only horses [Hippidion and Equus (Amerhippus)] and the giant sloth (Megatherium) are numerically significant (Politis et al., 1987, 1995). For the early component of this site, three radiocarbon dates were obtained on Equus (Amerhippus) neogeus (AA-7964, with AMS 11250±105 yrs BP; OXA-4590, with AMS 11000 ± 100 yrs BP and TO-1504, with AMS 8890±90 yrs BP). Tonni et al. (2003) also present radiocarbon data on Equus (Amerhippus) neogeus from Quequén Grande (the upper part of Guerrero member, Luján Formation) with values of 10290±130 yrs BP (LP- 1235). In addition, the datum of 8890 yrs BP provides exact chronological information with regard to the latest record, or taxon-data (Grayson, 1987) of Equus (Amerhippus) neogeus.
The influence of early hunter-gatherers on horses is difficult to quantify. Not enough information is available to evaluate the extent to which the pressure caused by hunting affected the late populations of these species. There is evidence that verifies that humans were present in the Pampean Region around 11,500 yrs B P and that they coexisted with the extinct large mammals for three thousand years or more. Politis et al. (1995) suggest that humans played a secondary role in the extinction of the native South American fauna, but had a greater effect on immigrant fauna such as horses and mastodons. Miotti and Salemme (1999) were able to verify evidence of the consumption of horses by the cut marks they found in fossil bone.
It has been difficult for proponents of the climatic hypothesis to isolate a single climatic cause for the extinctions. Climate may have provoked changes in flora communities and as consequence altered herbivores diets and caused heightened periods of competition (Graham and Lundelius, 1984). Also, a shortened growing season at the end of the Pleistocene would have created environmental stress for large mammals (Guthrie, 1984). Although horses may have been able to adapt to any one of these environmental perturbations, the combination of all of them at the same time may have been devastating for species that showed more selective dietary adaptations. Undoubtedly, pressures exerted by huntergatherers may have also added to the environmental stresses.

Conclusion

The objective of this study was to reconstruct the ancient diet and habitat preference of Hippidion devillei, Hippidion principale and Equus (Amerhippus) neogeus from the Pampean Region using carbon and oxygen isotopic composition of teeth and bones. Carbon isotope analyses reveal that these horses had different food adaptations. Due to their high crowned teeth, it has traditionally been thought that horses fed on abrasive grasses. However, carbon isotopic data from the Pampean Region indicate that these horses ranged from mixed feeders to more specialized C₃ grazers. Hippidion devillei from lower Pleistocene were principally C₃ grazers. Specimens of Hippidion principale and Equus (Amerhippus) neogeus from the middle Pleistocene had a C₃ to mixed C₃-C₄ diet, while specimens from latest Pleistocene were primarily C₃ grazers. These results are partially corroborated by data from the lower to middle Pleistocene from Bolivia (MacFadden and Shockey, 1997; Mac- Fadden, 2000). These authors indicate that horses from Tarija ranged from largely C₄ grazers (Equus insulatus) to principally mixed-feeders (Hippidion principale and Hippidion devillei ) which suggests that they may have partitioned their available food resources across a broad spectrum of pastures.
We believe that horses from middle Pleistocene exhibited opportunistic feeding strategies and consequently may have been adapted to diverse habitats, even though the majority of equids from the middle Pleistocene showed a preference for mixedfeeding. In contrast, populations from late Pleistocene appear to have adapted to a more selective diet, which restricted their habitat preference. A similar isotopic pattern was found for Stegomastodon platensis from the Pampean Region (Sánchez et al., 2003). The authors propose that gomphotheres were driven to extinction because they were specialized feeders, adapted to a kind of plant that may have disappeared during the Holocene (Sánchez et al., 2004).
Several nutritional hypotheses postulated for latest Pleistocene megamammal extinctions operate under the assumption that extinct taxa had specialized diets. The resource partitioning preference of Hippidion principale and Equus (Amerhippus) neogeus from latest Pleistocene in Pampean Region supports these hypotheses.

Acknowledgements

We wish to thank the curators of the Museo de La Plata (Argentina), and Museo Argentino de Ciencias Naturales "Bernardino Rivadavia" of Buenos Aires, who were in charge of the remains. The manuscript was greatly improved by thoughtful review from Gustavo Martínez and Gustavo Politis, and also by Martín Ubilla and an anonymous reviewer. Mr. James Watkins revised the English text. Samples were analyzed in the Isotope Geochemistry Laboratory at the Università di Trieste, and Servicio de Isótopos Estables of the Universidad Autónoma de Madrid (SIDI). This work was supported by a joint Research Project from the Agencia Española de Cooperación Iberoamericana, Spain - Argentina (2000-2001) and CSIC 2002; Projects PB94-0071, PB97- 1250, and BTE2001-1864 from the Dirección General de Investigación Científica y Técnica of Spain; grants from the Universidad Nacional del Centro, and the CONICET Project PIP- 2000-02773, Argentina.

References

Alberdi, M.T. and Prado, J.L. 1992. El registro de Hippidion Owen, 1869 y Equus (Amerhippus) Hoffstetter, 1950 (Mammalia, Perissodactyla) en América del Sur. Ameghiniana 29: 265-284.
Alberdi, M.T. and Prado, J.L. 1993. Review of the genus Hippidion Owen, 1869 (Mammalia; Perissodactyla) from the Pleistocene of South America. Zoological Journal of the Linnean Society 108: 1-22.
Alberdi, M.T., Prado, J.L. and Ortiz Jaureguizar, E. 1995. Patterns of body size changes in fossil and living Equini (Perissodactyla). Biological Journal of the Linnean Society 54: 349-370.
Alroy, J. 2001. A Multispecies Overkill Simulation of the End- Pleistocene Megafaunal Mass Extinction. Science 292: 1893- 1896.
Ameghino, F. 1888. Rápidas diagnosis de algunos mamíferos fósiles nuevos de la República Argentina. Obras Completas, Buenos Aires V, pp. 469-480.
Borrero, L. 1984. Pleistocene extinctions in South America. Quaternary of South America and Antarctic Peninsula 2: 115-125.
Bryant, J.D. and Froelich, P.N. 1995. A model of oxygen isotope fractionation in body water of large mammals. Geochimica et Cosmochimica Acta 59: 4523-4537.
Bryant, J.D., Luz, B. and Froelich, P.N. 1994. Oxygen isotopic composition of fossil horse tooth phosphate as a record of continental paleoclimate. Palaeogeography, Palaeoclimatology, Palaeoecology 107: 303-316.
Cerling, T.E., Wang, Y., and Quade, J. 1993. Expansion of C4 ecosystems as an indicator of global ecological change in the Late Miocene. Nature 361: 344-45.
Cerling, T.E., Harris, J.M., MacFadden, B.J., Leakey, M.G., Quade, J., Eisenmann, V., and Ehleringer, J.R. 1997. Global vegetation change through the Miocene/Pliocene boundary. Nature 389: 153-158.
Cione, L.A. and Tonni, E.P. 1995. Chronostratigraphy and "Land- Mammal Ages" in the Cenozoic of Southern South America: Principles, practices, and the "Uquian" Problem. Journal of Paleontology 69: 135-159.
Cione, L.A. and Tonni, E.P. 1999. Biostratigraphy and chronological scale of upper-most Cenozoic in the Pampean Area, Argentina. Quaternary of South America and Antarctica Peninsula 12: 23-51.
Cione, A.L., Tonni, E.P. and Soibelzon, L. 2003. The Broken Zig- Zag: Late Cenozoic large mammal and tortoise extinction in South America. Revista del Museo Argentino de Ciencias Naturales 5: 1-19.
D'Angela, D. and Longinelli, A. 1990. Oxygen isotopic composition of fossil mammal bones of Holocene age: paleoclimatological considerations. Chemical Geology (Isotopic Geoscience Section) 103: 171-179.
Dansgaard, W. 1964. Stable isotopes in precipitation. Tellus 16(4): 436-468.
Delgado, A., Iacumin, P., Stenni, B., Sanchez, B., and Longinelli, A. 1995. Oxygen isotope variations of phosphate in mammalian bone and tooth enamel. Geochimica et Cosmochimica Acta 59: 4299-4305.
De Niro, M. J. and Epstein, J. 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochima et Cosmochimica Acta 42: 495-506.
Dillon, C.A. and Rabassa, J. 1985. Miembro La Chumbiada, Formación Luján (Pleistoceno, provincia de Buenos Aires): Una nueva unidad estratigráfica del valle del Río Salado. Primeras Jornadas Geológicas Bonaerenses, Resúmenes 27.
Ehleringer, J.R., Field, C.B., Lin, Z.F., and Kuo, C.Y. 1986. Leaf carbon isotope and mineral composition in subtropical plants along an irradiance cline. Oecologia 70: 520-526.
Ehleringer, J.R., Sage, R.F., Flanagan, L.B., and Pearcy, R.W. 1991. Climatic change and evolution of C4 photosynthesis. Trends in Ecology and Evolution 6: 95-99.
Fidalgo, F., De Francesco, F.O., and Colado, U. 1973. Geología superficial en las hojas Castelli, J.M. Cobo y Monasterio (Prov. de Buenos Aires). 5º Congreso Geológico Argentino (Buenos Aires), 4: 27-39.
Fidalgo, F., De Francesco, F.O., and Pascual, R. 1975. Geología superficial de la llanura Bonaerense. 6º Congreso Geológico Argentino (Bahía Blanca), Relatorio: 100-138.
Fidalgo, F., Riggi, J.C., Correa, H. and Porro, N. 1991. Los 'Sedimentos Postpampeanos' continentales en el ámbito bonaerense. Revista de la Asociación Geológica Argentina 47: 239-256.
Gervais, P. 1855. Recherches sur les Mammifères fossiles de l'Amérique méridionale. P. Bertrend, Libraire-Editeur, Paris, 63 pp.
Graham, R.W., and Lundelius, E.L. 1984. Coevolutionary disequilibrium and Pleistocene Extinction. In: P.S. Martin and R.G. Klein (eds.), Quaternary Extinctions: A Prehistoric Revolution. University of Arizona Press, Tucson, pp. 223-249.
Grayson, D.K. 1987. An analysis of the Chronology of Late Pleistocene Mammalian Extinctions in North America. Quaternary Research 28: 281-289.
Guthrie, R.D. 1984. Mosaics, Allelochemics and Nutrients. An Ecological Theory of Late Pleistocene Megafaunal extinction. In: P.S. Martin and R.G. Klein (eds.), Quaternary Extinctions: A Prehistoric Revolution. University of Arizona Press, Tucson, pp. 259-298.
Iriondo, M.H. and García, N.O. 1993. Climatic variations in the Argentine plains during the last 18,000 years. Palaeogeography, Palaeoclimatology, Palaeoecology 101: 209-220.
King, J. E. and Saunders, J. J. 1984. Environmental insularity and the extinction of the American mastodont. In: P.S. Martin and R.G. Klein (eds.), Quaternary Extinctions: A Prehistoric Revolution. University of Arizona Press, Tucson, pp. 315-339.
Koch, P.L., Fisher, D.C., and Dettman, D. 1989. Oxygen isotope variation in the tusks of extinct proboscideans: a measure of season of death and seasonality. Geology 17: 515-519.
Koch, P.L., Behrensmeyer, A.K., Tuross, N. and Fogel, M.L. 1990. The fidelity of isotopic preservation during bone weathering and burial. Annual Report of the Director Geophysical Laboratory, Carnegie Institution of Washington 1989-1990: 105-110.
Koch, P.L., Fogel, M.L. and Tuross, N. 1994. Tracing the diets of fossil animals using stable isotopes. In: K. Lajtha and R.H. Michener (eds.), Stable isotopes in Ecology and environmental Science. Blackwell Scientific, pp. 63-92
Koch, P.L., Tuross, N., and Fogel, M.L. 1997. The effects of sample treatment and diagenesis on the isotopic integrity of carbon in biogenic hydroxylapatite. Journal of Archaeological Science 24: 417-429.
Kohn, M.J. 1996. Predicting animal ?18O: accounting for diet and physiological adaptation. Geochimica et Cosmochimica, Acta 60: 4811-4829.
Kohn, M.J., Schoeninger, M.J., and Valley, J.W. 1996. Herbivore tooth oxygen isotope compositions: effects of diet and physiology. Geochimica et Cosmochimica, Acta 60: 3889-3896.
Kohn, M.J., Schoeninger, M.J., and Valley, J.W. 1998. Variability in oxygen isotope compositions of herbivore teeth: reflections of seasonality or developmental physiology? Chemical Geology 152: 97-112.
Kolodny, Y., Luz, B. and Navon, O. 1983. Oxygen isotope variations in phosphate of biogenic apatites, I. Fish bone apatite. Rechecking the rules of the game. Earth and Planetary Science Letters 64: 398-404.
Lee-Thorp, J.A. and van der Merwe, N.J. 1987. Carbon isotope analysis of fossil bones apatite. South African Journal of Science 83: 712-715.
Lee-Thorp, J.A., van der Merwe, N.J., and Brain, C.K. 1989. Isotopic evidence for dietary differences between two extinct baboon species from Swartkrans. Journal of Human Evolution 18: 183-190.
Lee-Thorp, J.A., van der Merwe, N.J., and Brain, C.K. 1994. Diet of Australopithecus robustus at Swartkrans from stable carbon isotopic analysis. Journal of Human Evolution 27: 361-372.
Lessa, E.P., and Fariñas, R.A. 1996. Reassessment of extinction patterns among the Late Pleistocene mammals of South America. Palaeontology 39: 651-662.
Lessa, E.P., Valkenburgh, B.van and Fariñas, R.A. 1997. Testing hypotheses of differential mammalian extinctions subsequent to the Great American Biotic Interchange. Palaeogeography, Palaeoclimatology, Palaeoecology 135: 157-162.
Longinelli, A. 1984. Oxygen isotopes in mammal bone phosphate: a new tool for paleohydrological and paleoclimatological research? Geochimica et Cosmochimica, Acta 48: 385-390.
Longinelli, A. and Nuti, S. 1973. Oxygen isotope measurements of phosphate from fish teeth and bones. Earth and Planetary Science Letters 20: 337-340.
Lund, P.W. 1840. Blik paa Brasiliens Dyreverden for Sidste Jordomvaeltning. Tredie Afhandling: Forsaettelse af Pattedyrene. Det Kongelige Danske Videnskabernes Selskbas Naturvidenskabelige og Mathematiske Afhandlinger 8: 217-272.
Lund, P.W. 1846. Meddlelse af det Udbytte de I 1844 undersögte Knoglehuler Have avgivet til hundskaben om Brasiliens Dyreverden för sidste Jordomvaeltning. Det köngelige Danske Videnskabernes Selskabs naturvidenskabelige og mathematisk Afhandlinger 12, 57-94.
Luz, B., Kolodny, Y. and Horowitz, M. 1984. Fractionation of oxygen isotopes between mammalian bone-phosphate and environmental drinking water. Geochimica et Cosmochimica, Acta 48: 1689-1693.
MacFadden, B.J. 2000. Middle Pleistocene climate change recorded in fossil mammal teeth from Tarija, Bolivia, and upper limit of the Ensenadan Land-Mammal Age. Quaternary Research 54: 121-131.
MacFadden, B.J., Cerling, T.E., Harris, J.M., and Prado, J.L. 1999. Ancient latitudinal gradients of C₃/C₄ grasses interpreted from stable isotopes of New World Pleistocene horses. Global Ecology and Biogeography 8: 137-149.
MacFadden, B.J., Cerling, T.E. and Prado, J. L. 1996. Cenozoic Terrestrial Ecosystem in Argentina Evidence from Carbon isotopes of Fossil Mammal Teeth. Palaios 11: 319-327.
MacFadden, B.J., and Shockey, B.J. 1997. Ancient feeding ecology and niche differentiation of Pleistocene mammalian herbivores from Tarija, Bolivia: morphological and isotopic evidence. Paleobiology 23: 77-100.
Martin, P.S. 1967. Prehistoric Overkill. In: P.S. Martin and H.E. Wright Jr. (eds.), Pleistocene Extinction. The Search for a Cause. Yale University Press, New Haven. pp. 75-120.
Martin, P.S. 1973. The discovery of America. Science 179: 969-974.
Martin, P.S. 1984. Prehistoric Overkill: The Global Model. In: P.S. Martin and R.G. Klein (eds.), Quaternary Extinctions: A Prehistoric Revolution. University of Arizona Press, Tucson. pp. 354-403.
Miotti, L. and Salemme, M. 1999. Biodiversity, taxonomic richness and specialists-generalists during Late Pleistocene/Early Holocene times in Pampa and Patagonia (Argentina, Southern South America). Quaternary International 53/54: 53-68.
Politis, G. and Gutiérrez, M. 1998. Gliptodontes y cazadoresrecolectores de la Región Pampeana (Argentina). Latin American Antiquity 9: 111-134.
Politis G.G. and Prado, J.L. 1990. La influencia del hombre en las extinciones faunísticas del Pleistoceno/Holoceno en América: el caso pampeano. Memorias del Simposio de Arqueología y Antropología Física del 5º Congreso Nacional de Antropología, Memoria de Eventos Científicos ICFES (Colombia): 31-63 pp.
Politis, G., Tonni, E.P., Fidalgo, F., Salemme, M. and Guzmán, L. 1987. Man and Pleistocene Megamammals in the Argentine Pampa: Site 2 at Arroyo Seco. Current Research in the Pleistocene 4: 159-161.
Politis, G.G., Prado, J.L. and Beukens, R.P. 1995. The Human Impact In Pleistocene-Holocene Extinctions. In: E. Johnson (ed.), South America-The Pampean Case. Ancient Peoples and Landscapes. Museum of Texas Tech University, Texas. pp. 187- 205.
Prado, J.L. and Alberdi, M.T. 1994. A quantitative review of the horse Equus from South America. Palaeontology 37: 459-481.
Prado, J.L., Alberdi, M.T., Azanza, B. and Sánchez, B. 2001a. Climate and changes in mammal diversity during the late Pleistocene-Holocene in the Pampean Region (Argentina). Acta Palaeontologica Polonica 46: 261-276.
Prado, J.L., Azanza, B., Alberdi, M.T. and Gómez, G. 2001b. Mammal community and Global Change during the Late Pleistocene-Holocene in the Pampean Region (Argentina). In: D. Büchner (ed.), Studia honoraria: Studien in Memorian Wilhelm Schüle 11: 362-375.
Quade, J., Cerling, T.E., Barry, J.C., Morgan, M.E., Pilbeam, D.R., Chivas, A. ., Lee-Thorp, J.A. and van der Merwe, N.J. 1992. A 16-Ma record of paleodiet using carbon and oxygen isotopes in fossil teeth from Pakistan. Chemical Geology (Isotope Geoscience Section) 94: 183-192.
Sánchez-Chillón, B. and Alberdi, M.T. 1996. Taphonomic modification of oxygen isotopic composition in some South American Quaternary mammal remains. In: G. Meléndez, M. F. Blasco, and I. Pérez (eds.), 2º Reunión de Tafonomía y Fosilización. Institución Fernando el Católico, Zaragoza, pp. 353-356.
Sánchez-Chillón, B., Alberdi, M.T., Leone, G., Bonadonna, F.P., Stenni, B. and Longinelli, A. 1994. Oxygen isotopic composition of fossil equid tooth and bone phosphate: an archive of difficult interpretation. Palaeogeography, Palaeoclimatology, Palaeoecology 107: 317-328.
Sánchez-Chillón, B., Prado, J.L. and Alberdi, M.T. 2003. Paleodiet, ecology, and extinction of Pleistocene gomphotheres (Proboscidea) from Pampean Region (Argentina). Coloquios de Paleontología, Volumen Extraordinario 1: 617-625.
Sánchez-Chillón, B., Prado, J.L. and Alberdi, M.T. 2004. Feeding ecology, dispersal, and extinction of South American Pleistocene gomphotheres (Gomphotheriidae, Proboscidea). Paleobiology 30: 146-161.
Schultz, P.H., Zárate, M., Hames, W., Camilión, C. and King, J. 1998. A 3.3 Ma impact in Argentina and possible consequences. Science 282: 2061-2063.
Smith, B.N. and Epstein, S. 1971. Two categories of 13C/12C ratios for higher plants. Plant Physiology 47: 380-384.
Sullivan, C.H. and Krueger, H.W. 1981. Carbon isotope analysis of separate chemical phases in modern and fossil bone. Nature 301: 177-178.
Teruggi, M.E. 1957. The nature and origin of Argentine loess. Journal of Sedimentary Petrology 27: 322-332.
Tonni, E.P., Huarte, R.A., Carbonari, J.E. and Figini, A.J. 2003. New radiocarbon chronology for the Guerrero Menber of the Luján Formation (Buenos Aires, Argentina): palaeoclimatic significance. Quaternary International 109-110: 45-48.
Vogel, J.C. 1978. Isotopic assessment of the dietary habitats of ungulates. South African Journal of Science 74: 298-301.
Vogel, J.C., Fuls, A. and Ellis, R.P. 1978. The geographical distribution of kranz grasses in South Africa. South African Journal of Science 74: 209-215.
Vrba , E.S.1993. Turnover-pulses, the Red Queen, and related topics. American Journal of Science 293A: 418-452.
Webb, S.D. 1991. Ecogeography and the Great American Interchange. Paleobiology 17: 266-280.
Zárate, M.A. and Blasi, A. 1993. Late Pleistocene-Holocene Eolian deposits of the southern Buenos Aires Province, Argentina. A preliminary model. Quaternary International 17: 15-20.        [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]         [ Links ]

Recibido: 23 de junio de 2004.
Aceptado: 1 de setiembre de 2005.